A Matter of Balance

Incognito continues with his "back to basics" focus. This time is the turn of the analytical balance.

It's happened again. Following hot on the heels of "Pipette-gate" at the end of last year,¹ I've been involved in another dispute regarding the validity of a basic laboratory skill. This time it's the turn of measurement of mass. Because my previous article on the use of pipettes sparked so much debate and interest, I thought this a good opportunity to start another global difference of opinion on the use of balances within the analytical laboratory.

Photo Credit: ElementalImaging/Getty Images

Starting with the same question as last time - "Who legislates on the correct use of balances in the analytical laboratory?" - ask yourself who wrote your standard operating procedure (SOP) on the use of analytical balances to determine mass (not weight!)? Which source was used to inform and guide the author to ensure best practice was taken into account as well as any regulatory guidelines? Is the SOP always followed?

There are several regulations or regulatory bodies that advise on the correct use of balances; however, those which I use as the de facto standards are USP Chapter 41 <Balances>,² USP Chapter 1251 <Weighing on an Analytical Balance>,³ and ISO 17025 Lab14 Calibration of Weighing Machines (Edition 4 November 2006).⁴ Those of you who follow these things will know that USP Chapter 41 changed in December 2013 (after some 20 years in its previous version), and that it is mandatory in the pharmaceutical industry when testing to USP standards. USP <1251> is a guideline, mainly dealing with installation qualification/operational qualification/performance qualification (IQ/OQ/PQ) matters and outside the scope of our discussion here.

Let me highlight some of the issues that I have seen in the determination of mass, one of which was the cause of this latest dispute on the correct use of an analytical balance. Unlike the previous article on pipetting I'm not going to formulate an imaginary "perfect" SOP for balance use. It should be reasonably straightforward to assemble a reasonable SOP from a reference source, even if this is the manufacturer's instruction manual. Instead I'm going to highlight issues that are often not covered in SOPs, common mistakes, and bad practice in the measurement of mass.

General Weighing Practices

Location: Balances should be located away from sources of heat (out of direct sunlight and away from radiators or air conditioning units) and in a position where temperature does not fluctuate considerably (typical temperature drift is 1–2 ppm/°C). You should know how to calculate the temperature coefficient for sensitivity of your balances should you ever need to apply a correction factor! The balance should be placed in the corner of a room to ensure the least vibration, and should be away from doors, windows, or air conditioning units that can cause air draughts. Ideally, balance tables should be made of stone (should at least not suffer from sag or vibration); be anti-magnetic (no steel in the construction); and anti-static (no plastic or glass). The balance should be level (check bubble indicator before each weighing session), and, if adjustment is necessary, the balance should be recalibrated (internal sensitivity adjustment) before use.

Standby versus Power Off: Note that if a balance has been powered off (rather than put into standby mode), it may take between one hour and one day to "acclimatize" and you should consult your balance literature for the recommended time. If in standby mode, the balance may be used once the internal calibration routine is complete (where appropriate), or once a stable reading is achieved with nothing on the balance pan. It is good practice when beginning a series of weighings to load the balance pan, unload it again, and then tare before beginning to avoid an "initial weighing effect".

Humidity: The humidity of the balance location should be between 40% and 60% relative humidity (RH). High humidity can lead to condensation and adsorption of water; low humidity risks increased electrostatic interference. If the balance reading constantly drifts in one direction or is not repeatable between weighings of the same sample, use an anti-static gun, metal receiver, or a balance with antistatic ionizing blowers. Also note that changes in humidity can affect air buoyancy (see later).

Sample Handling: The balance pan and its surroundings should be free from powders and liquids and should be brushed clean or dried and recalibrated if found in a dirty state.

Use the smallest weighing vessel possible to reduce the effect of flow forces, don't touch with bare fingers (skin oil residues can reduce weighing accuracy), and use a receiver with a smaller rather than a wider neck to help prevent evaporation. Avoid using plastic weighing vessels or receivers because they are more susceptible to electrostatic effects.

Ensure the sample is temperature equilibrated with the weighing environment. Samples that are too cold will register a higher mass and vice versa for warm samples. The readout may be unstable if samples are not thermally equilibrated and air buoyancy effects will be markedly increased. Warm samples may also suffer from evaporation.

If weighing hygroscopic materials, ensure that the container closure is in place and that the receiving vessel is as narrow as possible (for example, stoppered volumetric flask). Note that fingerprints are hygroscopic!

Place the weighing vessel in the centre of the balance pan, otherwise the result will be skewed because of a phenomenon that is known as "eccentricity" or "off-centre loading error".

Samples containing magnetic materials (nickel, iron, steel, magnetic stirrer bars!!) can affect the measurement of mass. In this case, you may need to use antimagnetic balance pans or increase the distance from the pan to the magnetic sample by using an upside down beaker. You will typically note that balance readings are stable but not repeatable for the same sample or when positioning the sample at different positions on the balance pan.

Balance Calibration

You should adjust the sensitivity of the balance daily (usually via the balance internal calibration algorithm) or when you operate the balance for the first time; when you change the location of the balance; after levelling the balance; and after major changes in temperature, humidity, or air pressure.

Sensitivity is defined as "change in the output variable of a measuring instrument divided by the associated change in the input variable". For a balance, this is the change in the weighing value (mass displayed) ΔW divided by the load variation Δm (mass measured). Sensitivity is one of the most important specifications of a balance and is typically determined from the slope of the curve of mass measured versus mass displayed over the nominal range of the balance. Most analytical balances have a built in calibration function that can assess the sensitivity and linearity of the instrument.

Adjustment for Gravity: The further a mass is from the centre of the earth, the lower the gravitational force action upon it. A mass of 200 g measured on the ground floor will be 199.99937 g on the 10th floor - an error of 3.15 ppm. The further a mass is from the equator, the lower the centrifugal acceleration because of the rotation of the earth that counteracts gravitational force (around 92 ppm per degree of latitude!). It is therefore important to calibrate balance sensitivity in the balance location - especially after moving the balance from one location to another.

For highly accurate weighing, or when comparative weighing is done on different days, changes in environmental factors must be considered. The sensitivity of the balance is calibrated using weights of 8.0 g/cm³ and when measuring the mass of substances with different densities, air buoyancy errors can arise. A full treatment of this subject is outside our scope here; however, changes in atmospheric pressure, atmospheric humidity, and temperature all require an adjustment to be made for air buoyancy. You can find a good reference source at this link: http://www.npl.co.uk/upload/pdf/buoycornote.pdf

Complying with USP General Chapter <41> Requirements

USP Chapter 41 "repeatability" defines the starting point of a balance operating range:

• Perform 10 measurements with the exact same weight. Use a single certified weight (see ISO 17025 below) below 5% of the balances nominal range.

• Calculate 2× standard deviation (SD)/nominal value - the result should be <0.10%.

• Calculate the starting point of the operating range: 2 × SD × 1000.

• If SD <0.41 d, replace it by 0.41 d.

The coefficient "d" is the measurement interval or "readability". Examples include:

• Microbalances: 1 d = 1 μg = 0.000001 g (6-digit) (0.41 d = 0.41 μg)

• Semi-microbalances: 1 d = 0.01 mg = 0.00001 g (5-digit) (0.41 d = 0.041 mg)

• Analytical balances: 1 d = 0.1 mg = 0.0001 g (4-digit) (0.41 d = 0. 41 mg)

So, if the standard deviation of the determination of 10 measurements on your analytical balance is <0.41 × d, then the minimum weighable amount is 2 × 0.41 × 0.1 × 1000 = 82 mg

This minimum weight should be periodically assessed and is a unique feature of each balance - not the balance type, model, or manufacturer.

Accuracy: Balance accuracy must be assessed using a test weight between 5% and 100% of the balance capacity, and the indicated reading should be less than or equal to 0.1% of the test weight nominal value. The test weight should have a maximum permissible error of 0.03%.

One really important provision in USP <41> states that: Unless otherwise specified, when substances must be "accurately weighed" the weighing shall be performed using a balance that is calibrated over the operating range and meets the requirements defined for repeatability and accuracy.

How do you calibrate your balance? Is it done by your external service provider or metrology department? Note that here we don't mean the daily adjustment of the balance that most folks carry out (internal instrument calibration) but a full assessment of repeatability, sensitivity, linearity, departure of indication from nominal value, and eccentric loading. Most folks will use the ISO17025 guidelines for balance calibration and the United Kingdom Accreditation Service (UKAS) document Lab 14 (http://www.ukas.com/library/Technical-Information/Pubs-Technical-Articles/Pubs-List/LAB14.pdf) that outlines the requirements for test weights and calibration procedures nicely.

I think that will probably do for this instalment, except to say that I haven't touched upon the estimation of measurement uncertainty when using an analytical balance or indeed the frequency with which checks should be carried out. I will leave this to your own further study, but would highlight that under the new USP <41> regulations, daily balance sensitivity calibration is not required and that you should be familiar with the "risk-based approach" that is described therein.

The argument that prompted this article was regarding the proper (simple) specification in our in-house balance SOP of a "safety factor", which should be considered when specifying the smallest net weight measured using a balance to avoid out-of-specification measurements as a result of fluctuations in the minimum weighable amount. Any suggestions?

References

1. Incognito, The Column 10(21), 2–5 (2014).

2. General Chapter <41> "Balances" in Second Supplement to USP36 – NF31 (United States Pharmacopeial Convention, Rockville, MD, USA, 1 June 2013).

3. General Chapter <1251>"Weighing on an Analytical Balance" in Second Supplement to USP36 – NF31 (United States Pharmacopeial Convention, Rockville, MD, USA, 1 June 2013).

4. http://www.ukas.com/library/Technical-Information/Pubs-Technical-Articles/Pubs-List/LAB14.pdf

Contact author: Incognito

E-mail: admin@chromatographyonline.com

This article is from LCGC's digital magazine, The Column. Click here to view the issue.